CVE-2026-91943— Crawl4AI before 0.9.3 SSRF via PDFContentScrapingStrategy
Quick assessment
Affected
unclecode crawl4ai
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.
Crawl4AI 在 0.9.3 版本之前存在一个服务器端请求伪造（SSRF）漏洞，位于 中。该漏洞的成因是 方法使用 Python 库重新下载目标文件，但缺少出口（egress）验证。经过身份验证的攻击者可以提供一个重定向到内部地址的 URL，或利用 DNS 重绑定技术访问内部服务，并通过爬取结果中的 PDF 文本提取功能将响应内容泄露出去。
Possible ATT&CK Techniques 1 AI
T1557 · Adversary-in-the-Middle
Get alerts for future matching vulnerabilities
Log in to subscribe
I. Basic Information for CVE-2026-91943
Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗
Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
Vulnerability Title
Crawl4AI before 0.9.3 SSRF via PDFContentScrapingStrategy
Source: CVE Program / CVE List V5
Vulnerability Description
Crawl4AI before 0.9.3 contains a server-side request forgery vulnerability in PDFContentScrapingStrategy where _get_pdf_path() re-downloads targets with Python requests without egress validation. Authenticated attackers can supply URLs that redirect to internal addresses or use DNS rebinding to access internal services, exfiltrating responses through PDF text extraction in crawl results.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N
Source: CVE Program / CVE List V5
Vulnerability Type
服务端请求伪造(SSRF)
Source: CVE Program / CVE List V5
Affected Products
II. Public POCs for CVE-2026-91943
| # | POC Description | Source Link | Shenlong Link |
|---|
AI-Generated POC
Premium
No public POC found.
Login to generate AI POCIII. Intelligence Information for CVE-2026-91943
请登录查看更多情报信息。
Vendor Advisories for CVE-2026-91943 (2)
-
-
🔗 Crawl4AI before 0.9.3 SSRF via PDFContentScrapingStrategy | Advisories | VulnCheck third-party-advisory
Same Patch Batch · unclecode · 2026-09-15 · 5 CVEs total
| CVE-2026-91941 | 7.5 HIGH | Crawl4AI before 0.9.3 Denial of Service via PDFContentScrapingStrategy |
| CVE-2026-91940 | 7.5 HIGH | crawl4ai before 0.9.3 Arbitrary File Write via PDFContentScrapingStrategy |
| CVE-2026-91944 | 6.1 MEDIUM | crawl4ai before 0.9.3 DOM-based XSS via Playground UI |
| CVE-2026-91942 | 5.4 MEDIUM | crawl4ai before 0.9.3 Cross-Site Scripting via innerHTML |
IV. Related Vulnerabilities
Same product: crawl4ai
-
CVE-2026-91944
crawl4ai before 0.9.3 DOM-based XSS via Playground UI -
CVE-2026-91942
crawl4ai before 0.9.3 Cross-Site Scripting via innerHTML -
CVE-2026-91941
Crawl4AI before 0.9.3 Denial of Service via PDFContentScrapi -
CVE-2026-91940
crawl4ai before 0.9.3 Arbitrary File Write via PDFContentScr -
CVE-2026-56260
Crawl4AI - Arbitrary File Write via output_path Parameter
Same vendor: unclecode
-
CVE-2026-91944
crawl4ai before 0.9.3 DOM-based XSS via Playground UI -
CVE-2026-91942
crawl4ai before 0.9.3 Cross-Site Scripting via innerHTML -
CVE-2026-91941
Crawl4AI before 0.9.3 Denial of Service via PDFContentScrapi -
CVE-2026-91940
crawl4ai before 0.9.3 Arbitrary File Write via PDFContentScr -
CVE-2026-57572
Crawl4AI: Unauthenticated RCE via Chromium launch-argument i
Same weakness: CWE-918
-
CVE-2026-19504
Fabric.js loadFromJSON Server-Side Request Forgery Vulnerabi -
CVE-2026-58485
mcp-searxng: DNS-resolved Private Hostname SSRF in `web_url_ -
CVE-2026-54688
mcp-searxng: SSRF in web_url_read: the internal-address guar -
CVE-2026-54549
Meta Ads MCP: Server-Side Request Forgery (SSRF) in `upload_ -
CVE-2026-58201
Lokka: Azure Resource Manager URL path validation issue
V. Comments for CVE-2026-91943
No comments yet